Datos generales

This section includes the characterisation of the calcined forms of STA-7 for their applications in adsorption, diffusion and catalysis.

3.3.3.1 Characterisation of Calcined SAPO(20) STA7 Material TGA

TGA profiles are influenced by the topology of the sample rather than the organic species present therein.6 STA-7 possesses a three-dimensionally connected pore system, connected via 8MR windows. As a consequence, the rate of loss of organic species upon calcination is swift, with all organic species being removed in a single event with a weight lost of ca. 17% (template content per unit cell of ca. 90%), despite there being two different organic species present (Fig. 3.37).

Fig. 3.37Typical TGA thermogram of SAPO(20) STA-7.

The TGA profile shows a weight lost of ca. 17% related to the removed organic. That means a 90% of organic content per unit cell. The optimum calcination conditions to remove the organic species to obtain the free void volume are 550 ºC in a stream of dry

oxygen (Fig. 3.38). These conditions were used for experimental calcinations, where the temperature was increased from room temperature to the desired maximum at 1.5 ºC/min and held at the final temperature for 12 hours. The slow rate ramp and long dwell time was used to ensure all template was removed. Once the templates were removed, X-ray powder diffraction was conducted to check crystallinity (see Fig. 3.3 for the XRD patterns).

Fig. 3.38Free void space in STA-7 framework upon calcination, as shown by A. Fecant (IFP-Lyon).

SXRD and Refinement

Previously, the as-prepared structure of SAPO(20) STA-7 has been discussed in detail. It is also important to observe the structure once the material is calcined, because it is the calcined sample that possesses adsorptive and catalytic properties. For the first time SXRD was performed on a calcined SAPO STA-7 crystal (Table 3.14).

Rietveld refinement of the calcined structure against laboratory powder data was performed with the GSAS program suite using the atomic coordinates determined by SXRD of the as-prepared material (given in Table A.1, Appendix A) as a starting model. Instrumental parameters (background -type 2-, zero point, peak profile coefficients (pseudo-Voigtian peak shape) and structural parameters (unit cell, atomic coordinates, displacement parameters) were refined, constraining the displacement parameters to a single value, which refined to 0.0020(2) Å2 and restraining bond lengths, in order to maintain a chemically reasonable framework geometry. The final fit was achieved with

Rwp = 7.8% and Rp = 5.4%, with regions between 12.6 – 13.0, 15.9 – 16.1 and 20.5 – 20.8º 2θ excluded due to minor AlPO-18 (AEI) intergrowth at the surface of the crystals (the refined atomic coordinates and thermal parameters as well as bond lengths and angles are provided in Tables B.1, B.2 and B.3 in Appendix B). Figure 3.39 shows the plot of this refinement.

The refined unit cell parameters obtained area = b= 18.6931(7) Å and c= 9.4191(5) Å in P4/n and bond lengths P-O= 1.53(6) Å and Al-O = 1.74(7) Å (the unit cell values are higher than those observed by SXRD analysis, see Table 3.14 for a comparison of the unit cell values in the as-prepared and calcined using the different structural methods available. For the unit cell refinement, Rietveld method is more accurate than SXRD due to the geometry of the instrument).

Table 3.14Unit cell parameters for SAPO(20) STA-7 in as-prepared and calcined from different structural methods (*STOE software, **GSAS software, space group P4/n)).

Structural Method a=b(Å) c(Å)

SXRD (as-prepared) 18.656(15) 9.378(7) SXRD (calcined hydrated) 18.518(7) 9.246(4) PXRD* (as-prepared) 18.7347(19) 9.3849(14) Refinement** (calcined dehydrated) 18.6931(7) 9.4191(5)

Fig. 3.39(a) Rietveld refinement plot of calcined SAPO(20) STA-7 (red: experimental data, green: simulated pattern, purple: difference plot), (b) expanded high angle region.

b) a)

Solid State NMR of calcined SAPO

Solid-state NMR was performed on calcined SAPO(10), SAPO(20) and SAPO(30) STA- 7 to study the silicon and aluminium coordination in the framework. The crystallinity of calcined SAPO materials depends strongly on the uptake of water, because the materials suffer hydrolysis after a long period of hydration. This is an important consideration for sample storage, delivery and activation. That issue is reported in literature for the case of SAPO-34 (CHA) where long-term hydration of a calcined sample over two years produces a loss of crystallinity by breaking Si-OH-Al bonds.19

The 29Si MAS NMR (DP) spectra of calcined samples lose resolution compared with those of as-prepared materials.20For SAPO(10) and SAPO(20), the spectra show a broad resonance, possibly with more than one peak corresponding to different crystallographic sites, at –94.8 and –95.2 ppm respectively, attributed to Si(4Al) as expected.21At higher silicon contents the silicate islands resonance that is poorly defined in the as- prepared material is more obvious in the spectrum of the calcined solid, with an extra signal at – 108.6 ppm related in literature to Si(4Si).4 The main signal at –89.4 ppm is in the region related to Si(4Al) and Si(3Al,1Si), although it is rather broad and signal related to Si(2Al,2Si) could also be present (Fig. 3.40).

Fig. 3.4029Si MAS NMR spectra of SAPO(10), SAPO(20) and SAPO(30) STA-7(a, b and c1 respectively, c2 is the deconvoluted curves of c1 spectrum).

-120 -110 -100 -90 -80 -70 -60 ppm a) -120 -110 -100 -90 -80 -70 -60 ppm b) c2) ppm ppm ppm ppm -100 -50 c1)

The 27Al MAS NMR spectrum of sample SAPO(30) shows an acute decrease of the signal related to aluminium in tetrahedral position at 33.7 ppm and in contrast a sharper signal at –11.6 ppm arises. This extra signal is attributed to 6-coordinated Al in the literature;22 probably Al is coordinated to water molecules from moisture, leading to partial removal of aluminum from the framework. As a result, 31P MAS NMR shows an extra shoulder, probably due to Al-O-P broken bonds and the29Si MAS NMR shows the formation of silica islands in a more predominant manner than in the as-prepared form. The XRD pattern of the SAPO(30) STA-7 after one year after calcination does not show any remarkable loss of crystallinity (Fig. 3.41). Therefore, SAPO materials once calcined remain stable although moisture can cause local redistribution of the framework atoms.

Fig. 3.41 (left) 27Al and31P MAS NMR spectra of SAPO(30) STA-7 in the as-prepared and calcined form (top and bottom, respectively).(right) XRD patterns of SAPO(30) STA-7 in the as-prepared, freshly calcined and taken after one year of calcination (from top to bottom).

2Theta 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 0.0 20.0 40.0 60.0 80.0 100.0 120.0 140.0 160.0 180.0 R e la ti v e In te n s it y (% )

MC125(g) SAPO(30) STA-7 CALC and runned after min of one year / Range 1

MC125(g)CALC 30%Si in Gel / Range 1

MC125(g) SAPO STA-7 30%Si in Gel / Range 1

SAPO(30) STA-7 one year after calcination SAPO(30) STA-7 as-prepared

SAPO(30) STA-7 freshly calcined

-50 0 ppm -100 0 100 ppm -50 0 ppm -100 0 100 ppm

N2adsorption

Nitrogen adsorption was conducted gravimetrically on a calcined sample of SAPO(20) STA-7 at 77 K. The resulting isotherm exhibits a typical type I profile. A maximum uptake of 25% by mass suggests an internal pore volume of 0.29 cm3/g, assuming adsorbed N2 to have a density similar to that of liquid nitrogen at 77 K (0.81 g/cm3), see

figure 3.42. The adsorption properties of this and other related materials obtained from this work will be discussed in detail in part three.

0 5 10 15 20 25 30 -0.04 0.06 0.16 0.26 0.36 0.46

Relative Pressure (p/po)

M a s s U p ta k e (% ) Adsorption Desorption

Fig. 3.42Nitrogen adsorption/desorption isotherm for calcined SAPO(20) STA-7.

FTIR

FTIR spectroscopy provides information about the bridging hydroxyl groups (OH) formed upon calcination. The introduction of silicon into the neutral AlPO forms a negatively charged framework that is balanced by the positive template in the as-prepared material. During calcination the negative charge is balanced by protons attached to bridging oxygen atoms from the T positions where phosphorus atoms have been substituted by silicon atoms. These bridging hydroxyl groups are the Brønsted acid sites available for catalytic applications.

The following spectrum shows two bands (Fig. 3.43): a sharp and intense one between 3600-3625 cm-1and a broader one between 2900-2800 cm-1. The first one is related to the hydroxyl bridging groups (O-H stretching vibrations)23. Probably this signal is a

contribution of two bands related to two different Brønsted sites as reported in literature that are not fully resolved in the spectra.24The second band is related to water molecules adsorbed from moisture. This band decreases upon heating the sample, so that between 250 ºC and 350 ºC most of the water is removed and the signal disappears. For that reason, 300 ºC was the temperature used to activate the materials for adsorption and diffusion experiments.

Fig. 3.43FTIR spectrum of SAPO(20) STA-7 (courtesy of Prof. Jiri Cejka, J. Heyrovsky Institute).

Once the existence of the hydroxyl bridging groups was confirmed, d3-acetonitrile (CD3CN) was adsorbed to observe their acidity because acetonitrile acts as a base, interacting via the nitrogen lone pair. From the following spectra (Fig. 3.44), it can be observed that acetonitrile is indeed adsorbed, which resulted in a substantial decrease in the hydroxyl group band intensity. Two new bands appear at 2258 and 2325 cm-1, due to free and chemisorbed acetonitrile respectively.25

3800 3600 3400 3200 3000 2800 2600 1,0 1,5 2,0 A b s o rb a n c e Wavenumber, cm-1 100°C 200°C 250°C 300°C 350°C 450°C 530°C MC3 SAPO STA-7 aktivace 060531, f=0,6022

4000 3800 3600 3400 3200 AA BA A b s o rb a n c e Wavenumber, cm-1 BA AA 2400 2350 2300 2250 2200 A b s o rb a n c e Wavenumber, cm-1

Fig. 3.44FTIR spectra from d3-acetonitrile adsorption in SAPO(20) STA-7, BA (before adsorption) and AA

(after adsorption) (courtesy of Professor Jiri Cejka, J. Heyrovsky Institute, Prague). 3.3.3.2 Porosity of Related STA-7 Materials

SAPO(20) STA-7 with extra framework cations by post synthesis treatment

Within the INDENS project, not only was the role of framework composition on adsorption to be studied, but also the interaction of gas molecules with extra framework cations. SAPO(20) STA-7 was used as a starting material for further attempts to introduce extra framework cations via two procedures:

a) in situ, as previously performed by R. Garcia and S. Warrender.5,6 The use of azamacrocycle metal complexes as templates results, once calcined, in extra framework cations within the pores. Cation inclusion is achieved in a convenient two-step process rather than a three-step sequence of synthesis, calcinations and subsequent ion exchange, which itself may not be entirely efficient, or may result in structural degradation.

In this work the effect on STA-7 crystallisation using Cu-cyclam and the new order of addition of reagents was studied. The purple as-prepared Cu-SAPO STA-7 became turquoise upon calcination indicating that the template has been removed leaving Cu2+in the pores. In addition, the new synthetic method gives crystals of different morphology from those observed previously. Whereas previous preparations show clear evidence of

intergrowth crystals, the new method produces tetragonal prismatic crystals. These appear to be an aggregation of layers, particularly at the surface. If tetragonal prisms of SAPO(20) STA-7 were used as seeds in the synthesis, the morphology changes to form large stratified cubic crystals (up to 50 μm) and tiny well shaped cubes (Fig. 3.45). Copper content in the larger crystals is proved by EDX analysis.

Fig. 3.45 SEM images of Cu-SAPO(20) STA-7: using previous synthetic methodology (a), via the new methodology (b, d) and incorporating SAPO(20) STA-7 crystals as seeds (c, e). EDX of the crystals showed in e) indicates the presence of copper.

b) Post synthesis ion exchange. In this approach the ability to ion exchange calcined SAPO STA-7 was measured. Ion exchange was performed using aqueous nitrate solutions. As a starting point 1M solutions (50 ml) of alkali metal cations (K+, Cs+) and earth metal cations (Mg2+, Ca2+) were used for 0.5 g of STA-7 sample. The exchange procedure was repeated two or three times (each time for about 2 hours) at 70 ºC under stirring conditions. a) d) 50μm e) b) c)

SEM showed significant change in the morphology of the crystals after ion exchange and the XRD patterns show a loss of crystallinity (Fig 3.46 and 3.47). ICP-AES (inductively coupled plasma atomic emission spectroscopy) reveal the limited ion exchange capacity (ca. 10%) for divalent cations and the more significant cation exchange for monovalent cations (ca. 50%), see Table 3.15. It may be that divalent cations are not readily able to enter the structure because of the large size of their hydrated ions.

Fig. 3.46SEM images of SAPO(20) STA-7 as made (a), calcined (b), after K+(c), Mg2+(d) and Ca2+(e) ion- exchange.

Table 3.15ICP-AES analysis on SAPO(20) STA-7 ion exchanged

Sample Al% Si% P% Cation% Exchanged

K-SAPO(20) STA-7 15.3 4.6 13.8 2.1 ~ 50% Ca-SAPO(20) STA-7 16.3 5.3 14.0 0.5 ~ 10% Mg-SAPO(20) STA-7 16.0 4.6 14.2 0.3 ~ 7%

c) d) e)

Fig. 3.47XRD patterns of SAPO(20) STA-7 calcined (blue), after K+(purple), Mg2+(cyan) and Ca2+(green) ion-exchange.

The main cause of the loss of crystallinity is likely to be due to slow hydrolysis of the framework. Therefore, several attempts were made, changing experiment time, number of repetitions, temperature and alkali metal cation sources. For K+ the best results were obtained using 1M KCl, repeating the procedure twice (each time for about 1 hour) heating at 60 ºC and stirring at 375 rpm; for Cs+, the procedure was followed only once, using the same conditions as K+but with 0.4M nitrate solution.

For these conditions, the XRD patterns indicate that the material remained crystalline and SEM images show no significant change in the crystal morphology (Fig. 3.48 and 3.49). EDX shows 63% and 31% ion exchange for K+ and Cs+ respectively under these conditions assuming 100% exchange corresponded to a positive charge for every Si atom in the framework. SXRD analysis was possible for the Cs+ form of SAPO STA-7. Although the data quality was poor (Rint = 0.26) it indicated the inclusion of Cs+. The cation can be located in the 8MR window of the large cage of the STA-7 structure with a Cs-O bond of ca. 3.3 Å (Fig. 3.50). EDX analysis suggests that the level of Cs+ is one third of the maximum calculated on the silicon level of the STA-7, but SXRD locates

2Theta 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 0.0 20.0 40.0 60.0 80.0 100.0 120.0 140.0 160.0 180.0 R e la ti v e In te n s it y (% )

MC109(g)CALCINED SAPO STA-7 20%Si in Gel / Range 1

MC109(g)K+2times / Range 1

MC110(g)Mg+2 / Range 1

MC110(g)Ca+2 / Range 1

Mg2+-SAPO(20) STA-7

K+-SAPO(20) STA-7

SAPO(20) STA-7 calcined

only around a quarter of this (occupancy of 0.0625 in general position). The atomic coordinates and related SXRD details are in Tables A.5, A.6, A.7 and A.8, Appendix A.

Fig. 3.48 XRD patterns of SAPO(20) STA-7 as-prepared (blue), calcined (purple), for K+(cyan) and Cs+ (green) ion exchange.

Fig. 3.49SEM image of SAPO STA-7(20) after K+(top) and Cs+(bottom) ion exchange.

2Theta 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 0.0 20.0 40.0 60.0 80.0 100.0 120.0 140.0 R e la ti v e In te n s it y (% ) MC63'afterCALCINATION / Range 1 MC63'(g)K+ / Range 1

MC122(g)Cs+ ion Exchange 1 time 0.05gsample 100ml CsNO3 0.4M 60C 1h / Range 1

SAPO(20) STA-7 as-prep. SAPO(20) STA-7 calcined

Cs+_{-SAPO(20) STA-7}

Fig. 3.50Coordinated environment of Cs+in SAPO(20) STA-7.

Porosities of STA-7 materials

Table 3.17 summarises the values of the total volume uptake from N2 isotherms of each material here synthesise stable after calcination to show their suitability as adsorbents:

Table 3.16Summary of materials with potential adsorption properties.

N2Isotherms

STA-7

Sample Extraframework Cation

% by mass Pore Volume cm3/g SAPO(20) STA-7 24.43 0.30 SAPO(30) STA-7 14.67 0.18 SAPO(10) STA-7 18.95 0.23 SAPO(20) STA-7 K 14.89 0.18 SAPO(20) STA-7 Cu 18.63 0.23 CoAPO STA-7 17.45 0.22 MnSAPO STA-7 20.67 0.26 MgSAPO STA-7 22.14 0.27

These data show that the STA-7 materials prepared could be used for CO2adsorption. The most promising for this purpose are SAPO(20) STA-7, the reference material, and also the MgAPSOs and CoAPO. The samples with ion exchange have a lower pore volume due to reduction of free pore space by the inclusion of cations, but they have retained porosity and so are potential CO2adsorbents.

3.4 Summary

The hydrothermal synthesis of the aluminophosphate-based STA-7 material has been explored in detail and a new route to high quality single crystals suitable for further applications has been established.

Structural characterisation of the as-prepared material combined with modelling, elucidates the co-templating effect of the co-base, tetraethylammonium cation (TEA). The inclusion of SiO2 into the AlPO gel results in silicoaluminophosphate frameworks. The mechanism of incorporation can be monitored by29Si MAS NMR and also,27Al MQ MAS NMR. These show that silicon substitutes mainly by phosphorus, giving Si(4Al) and, as a consequence, Al(1Si,3P) sites for samples up to Si/Al ratio 0.27. At higher content, a small number of silica islands are formed.

The study of calcined SAPO STA-7 by27Al MAS NMR shows that this is stable, porous and acidic, but needs to be stored dry.

1. P. A. Wright, M. J. Maple, A. M. Z. Slawin, V. Patinec, R. A. Aitken, A. Welch, P. A. Cox,J. Chem. Soc., Dalton Trans.,2000, 1243.

2. V. Patinec, P. A. Wright, P. Lightfoot, R. A. Aitken, P. A. Cox,J. Chem. Soc.,Dalton Trans., 1999, 3909.

3. W. G. Song, J. F. Haw, J. B. Nicholas, C. S. Heneghan,J. Am. Chem. Soc., 2000,122, 10726; W. G. Song, J. F. Haw,J. Am. Chem. Soc., 2001,123, 4749.

4. J. Chen, P. A. Wright, J. M. Thomas, S. Natarajan, L. Marchese, S. M. Bradley, G. Sankar, C. R. A. Catlow, P. L. Gai-Boyes, R. P. Townsend, C. M. Lok, J. Phys. Chem., 1994,98,10216.

5. R. Garcia Salas,Synthesis and Characterisation of Aluminophosphate-based Materials Prepared with Nickel Complexes as Structure Directing Agents, PhD Thesis, University of St Andrews, 2003.

6. S. J. Warrender, Structure Direction in the Formation of Zeolitic Materials, PhD Thesis, University of St Andrews, 2007.

7. STOE Win XPOW (version 1.04), STOE & Cie GmbH, Hilpertstrasse, D-64296, Darmstadt, 1999.

8. A. C. Larson, R. B. von Dreele, Generalized Crystal Structure Analysis System, Los Almos National Laboratory, USA, 1988.

9. SHELXTL 6.14, G M Sheldrick, Bruker AXS, Madison, 2004.

11. A.K. Rappé, C.J. Casewit, K.S. Colwell, W.A. Goddard, W.M. Skiff,J. Amer. Chem. Soc., 1992,114,10024; C.J. Casewit, K.S. Colwell, A.K. Rappé,J. Am. Chem. Soc., 1992,

114, 10035; C.J. Casewit, K.S. Colwell, A.K. Rappé, J. Am. Chem. Soc., 1992, 114, 10046; A.K. Rappé, K.S. Colwell, C.J. Casewit,Inorg.Chem.,1993,32, 3438.

12. S. Oliver, A. Kuperman, A. Lough, G. A. Ozin,J. Mater. Chem., 1997,7, 807. 13. P. S. Wheatley, R. E. Morris,J. Solid State Chem., 2002,167, 267.

14. H. Hobson, K. P. Lillerud, ‘Verified Synthesis of Zeolitic Materials’, 2nd Edition, Elsevier, Amsterdam, 2001.

15. J. F. Desreux, E. Merciny, M. F. Loncin,Inorg. Chem., 1981,20, 987. 16. P. J. Barrie, J. Klinowski,J. Phys. Chem., 1989,93, 5972.

17. C. S. Blackwell, R. L. Patton,J. Phys. Chem., 1988,92, 3965.

18. T.-H. Chen, B.H. Wouters, P.J. Grobet, Colloids and Surfaces A: Physicochemical and Engineering Aspects, 1999,158, 145.

19. M. Briend, R. Vomscheid, M. J. Peltre, P. P. Man, D. Barthomeuf, J. Phys Chem. 1995,99, 8270.

20. H. J. Jung, C.-H. Shin, S. B. Hong,J. Phys. Chem. B,2005,109,20847. 21. C. S. Blackwell, R. L. Patton,J. Phys. Chem., 1988,92, 3965.

22. R. Vomscheid, M. Briend, M.J. Peltre, P.P. Man, D. Barthomeuf, J. Phys. Chem., 1994,98, 9614.

23. S. A. Zubkov, L. M. Kustov, V. B. Kazansky, l. Girnus, R. Fricke, J. Chem. Soc.

Faraday Trans.,1991,87, 897.

24. I Déroche, G. Maurin, P. Llewellyn, M. Castro, P.A. Wright, M. Bejblova, J. Cejka,

From Zeolites to Porous MOF Materials the 40th Anniversary of International Zeolite Conference, R. Xu, Z. Gao, J. Chen, W. Yan (Eds.), Elsevier, Amsterdam, 2008, 1660. 25. G.M. Pearce, S. Bordiga, Pers. Comm.

Synthesis & Structure

Characterisation

of Zeotypes

Chapter 4: STA-14

_________________________________

4.1 Introduction

Having prepared the SAPO form of STA-7 using a co-templating strategy arrived at via a trial and error approach, this strategy was developed in attempts to synthesise, as an